Add MINMAX suffix to filenames of GEMM/IGEMM/PPMM/DWCONV micro-kernels

PiperOrigin-RevId: 305607024
diff --git a/BUILD.bazel b/BUILD.bazel
index 34235ce..c26c359 100644
--- a/BUILD.bazel
+++ b/BUILD.bazel
@@ -94,45 +94,45 @@
     "src/f32-dwconv-spchw/3x3s2p1-scalar.c",
     "src/f32-dwconv-spchw/5x5p2-scalar.c",
     "src/f32-dwconv-spchw/5x5s2p2-scalar.c",
-    "src/f32-dwconv/gen/up1x25-scalar-acc2.c",
-    "src/f32-dwconv/gen/up1x25-scalar.c",
-    "src/f32-dwconv/gen/up1x4-scalar-acc2.c",
-    "src/f32-dwconv/gen/up1x4-scalar.c",
-    "src/f32-dwconv/gen/up1x9-scalar-acc2.c",
-    "src/f32-dwconv/gen/up1x9-scalar.c",
-    "src/f32-dwconv/gen/up2x25-scalar-acc2.c",
-    "src/f32-dwconv/gen/up2x25-scalar.c",
-    "src/f32-dwconv/gen/up2x4-scalar-acc2.c",
-    "src/f32-dwconv/gen/up2x4-scalar.c",
-    "src/f32-dwconv/gen/up2x9-scalar-acc2.c",
-    "src/f32-dwconv/gen/up2x9-scalar.c",
+    "src/f32-dwconv/gen/up1x25-minmax-scalar-acc2.c",
+    "src/f32-dwconv/gen/up1x25-minmax-scalar.c",
+    "src/f32-dwconv/gen/up1x4-minmax-scalar-acc2.c",
+    "src/f32-dwconv/gen/up1x4-minmax-scalar.c",
+    "src/f32-dwconv/gen/up1x9-minmax-scalar-acc2.c",
+    "src/f32-dwconv/gen/up1x9-minmax-scalar.c",
+    "src/f32-dwconv/gen/up2x25-minmax-scalar-acc2.c",
+    "src/f32-dwconv/gen/up2x25-minmax-scalar.c",
+    "src/f32-dwconv/gen/up2x4-minmax-scalar-acc2.c",
+    "src/f32-dwconv/gen/up2x4-minmax-scalar.c",
+    "src/f32-dwconv/gen/up2x9-minmax-scalar-acc2.c",
+    "src/f32-dwconv/gen/up2x9-minmax-scalar.c",
     "src/f32-gavgpool-spchw/scalar-x1.c",
     "src/f32-gavgpool/7p7x-scalar-c1.c",
     "src/f32-gavgpool/7x-scalar-c1.c",
-    "src/f32-gemm/gen-inc/1x4-scalar.c",
-    "src/f32-gemm/gen-inc/2x4-scalar.c",
-    "src/f32-gemm/gen-inc/4x4-scalar.c",
-    "src/f32-gemm/gen/1x4-scalar.c",
-    "src/f32-gemm/gen/2x4-scalar.c",
-    "src/f32-gemm/gen/4x2-scalar.c",
-    "src/f32-gemm/gen/4x4-scalar.c",
+    "src/f32-gemm/gen-inc/1x4inc-minmax-scalar.c",
+    "src/f32-gemm/gen-inc/2x4inc-minmax-scalar.c",
+    "src/f32-gemm/gen-inc/4x4inc-minmax-scalar.c",
+    "src/f32-gemm/gen/1x4-minmax-scalar.c",
+    "src/f32-gemm/gen/2x4-minmax-scalar.c",
+    "src/f32-gemm/gen/4x2-minmax-scalar.c",
+    "src/f32-gemm/gen/4x4-minmax-scalar.c",
     "src/f32-hswish/gen/scalar-x1.c",
     "src/f32-hswish/gen/scalar-x2.c",
     "src/f32-hswish/gen/scalar-x4.c",
     "src/f32-ibilinear/gen/scalar-c1.c",
     "src/f32-ibilinear/gen/scalar-c2.c",
     "src/f32-ibilinear/gen/scalar-c4.c",
-    "src/f32-igemm/gen/1x4-scalar.c",
-    "src/f32-igemm/gen/2x4-scalar.c",
-    "src/f32-igemm/gen/4x2-scalar.c",
-    "src/f32-igemm/gen/4x4-scalar.c",
+    "src/f32-igemm/gen/1x4-minmax-scalar.c",
+    "src/f32-igemm/gen/2x4-minmax-scalar.c",
+    "src/f32-igemm/gen/4x2-minmax-scalar.c",
+    "src/f32-igemm/gen/4x4-minmax-scalar.c",
     "src/f32-maxpool/9p8x-scalar-c1.c",
     "src/f32-pavgpool/9p8x-scalar-c1.c",
     "src/f32-pavgpool/9x-scalar-c1.c",
-    "src/f32-ppmm/gen/2x4-scalar.c",
-    "src/f32-ppmm/gen/3x3-scalar.c",
-    "src/f32-ppmm/gen/4x2-scalar.c",
-    "src/f32-ppmm/gen/4x4-scalar.c",
+    "src/f32-ppmm/gen/2x4-minmax-scalar.c",
+    "src/f32-ppmm/gen/3x3-minmax-scalar.c",
+    "src/f32-ppmm/gen/4x2-minmax-scalar.c",
+    "src/f32-ppmm/gen/4x4-minmax-scalar.c",
     "src/f32-prelu/gen/scalar-2x1.c",
     "src/f32-prelu/gen/scalar-2x4.c",
     "src/f32-raddstoreexpminusmax/gen/scalar-lut64-p2-x1.c",
@@ -256,34 +256,34 @@
     "src/f32-clamp/gen/wasm-x1.c",
     "src/f32-clamp/gen/wasm-x2.c",
     "src/f32-clamp/gen/wasm-x4.c",
-    "src/f32-dwconv/gen/up1x25-wasm-acc2.c",
-    "src/f32-dwconv/gen/up1x25-wasm.c",
-    "src/f32-dwconv/gen/up1x4-wasm-acc2.c",
-    "src/f32-dwconv/gen/up1x4-wasm.c",
-    "src/f32-dwconv/gen/up1x9-wasm-acc2.c",
-    "src/f32-dwconv/gen/up1x9-wasm.c",
-    "src/f32-dwconv/gen/up2x25-wasm-acc2.c",
-    "src/f32-dwconv/gen/up2x25-wasm.c",
-    "src/f32-dwconv/gen/up2x4-wasm-acc2.c",
-    "src/f32-dwconv/gen/up2x4-wasm.c",
-    "src/f32-dwconv/gen/up2x9-wasm-acc2.c",
-    "src/f32-dwconv/gen/up2x9-wasm.c",
+    "src/f32-dwconv/gen/up1x25-minmax-wasm-acc2.c",
+    "src/f32-dwconv/gen/up1x25-minmax-wasm.c",
+    "src/f32-dwconv/gen/up1x4-minmax-wasm-acc2.c",
+    "src/f32-dwconv/gen/up1x4-minmax-wasm.c",
+    "src/f32-dwconv/gen/up1x9-minmax-wasm-acc2.c",
+    "src/f32-dwconv/gen/up1x9-minmax-wasm.c",
+    "src/f32-dwconv/gen/up2x25-minmax-wasm-acc2.c",
+    "src/f32-dwconv/gen/up2x25-minmax-wasm.c",
+    "src/f32-dwconv/gen/up2x4-minmax-wasm-acc2.c",
+    "src/f32-dwconv/gen/up2x4-minmax-wasm.c",
+    "src/f32-dwconv/gen/up2x9-minmax-wasm-acc2.c",
+    "src/f32-dwconv/gen/up2x9-minmax-wasm.c",
     "src/f32-gavgpool/7p7x-wasm-c1.c",
     "src/f32-gavgpool/7x-wasm-c1.c",
-    "src/f32-gemm/gen-inc/1x4-wasm.c",
-    "src/f32-gemm/gen-inc/2x4-wasm.c",
-    "src/f32-gemm/gen-inc/4x4-wasm.c",
-    "src/f32-gemm/gen/1x4-wasm.c",
-    "src/f32-gemm/gen/2x4-wasm.c",
-    "src/f32-gemm/gen/4x2-wasm.c",
-    "src/f32-gemm/gen/4x4-wasm.c",
+    "src/f32-gemm/gen-inc/1x4inc-minmax-wasm.c",
+    "src/f32-gemm/gen-inc/2x4inc-minmax-wasm.c",
+    "src/f32-gemm/gen-inc/4x4inc-minmax-wasm.c",
+    "src/f32-gemm/gen/1x4-minmax-wasm.c",
+    "src/f32-gemm/gen/2x4-minmax-wasm.c",
+    "src/f32-gemm/gen/4x2-minmax-wasm.c",
+    "src/f32-gemm/gen/4x4-minmax-wasm.c",
     "src/f32-hswish/gen/wasm-x1.c",
     "src/f32-hswish/gen/wasm-x2.c",
     "src/f32-hswish/gen/wasm-x4.c",
-    "src/f32-igemm/gen/1x4-wasm.c",
-    "src/f32-igemm/gen/2x4-wasm.c",
-    "src/f32-igemm/gen/4x2-wasm.c",
-    "src/f32-igemm/gen/4x4-wasm.c",
+    "src/f32-igemm/gen/1x4-minmax-wasm.c",
+    "src/f32-igemm/gen/2x4-minmax-wasm.c",
+    "src/f32-igemm/gen/4x2-minmax-wasm.c",
+    "src/f32-igemm/gen/4x4-minmax-wasm.c",
     "src/f32-maxpool/9p8x-wasm-c1.c",
     "src/f32-pavgpool/9p8x-wasm-c1.c",
     "src/f32-pavgpool/9x-wasm-c1.c",
@@ -342,57 +342,57 @@
     "src/f32-avgpool/9x-psimd-c4.c",
     "src/f32-clamp/gen/psimd-x4.c",
     "src/f32-clamp/gen/psimd-x8.c",
-    "src/f32-dwconv/gen/up4x25-psimd-acc2.c",
-    "src/f32-dwconv/gen/up4x25-psimd.c",
-    "src/f32-dwconv/gen/up4x4-psimd-acc2.c",
-    "src/f32-dwconv/gen/up4x4-psimd.c",
-    "src/f32-dwconv/gen/up4x9-psimd-acc2.c",
-    "src/f32-dwconv/gen/up4x9-psimd.c",
-    "src/f32-dwconv/gen/up8x25-psimd-acc2.c",
-    "src/f32-dwconv/gen/up8x25-psimd.c",
-    "src/f32-dwconv/gen/up8x4-psimd-acc2.c",
-    "src/f32-dwconv/gen/up8x4-psimd.c",
-    "src/f32-dwconv/gen/up8x9-psimd-acc2.c",
-    "src/f32-dwconv/gen/up8x9-psimd.c",
+    "src/f32-dwconv/gen/up4x25-minmax-psimd-acc2.c",
+    "src/f32-dwconv/gen/up4x25-minmax-psimd.c",
+    "src/f32-dwconv/gen/up4x4-minmax-psimd-acc2.c",
+    "src/f32-dwconv/gen/up4x4-minmax-psimd.c",
+    "src/f32-dwconv/gen/up4x9-minmax-psimd-acc2.c",
+    "src/f32-dwconv/gen/up4x9-minmax-psimd.c",
+    "src/f32-dwconv/gen/up8x25-minmax-psimd-acc2.c",
+    "src/f32-dwconv/gen/up8x25-minmax-psimd.c",
+    "src/f32-dwconv/gen/up8x4-minmax-psimd-acc2.c",
+    "src/f32-dwconv/gen/up8x4-minmax-psimd.c",
+    "src/f32-dwconv/gen/up8x9-minmax-psimd-acc2.c",
+    "src/f32-dwconv/gen/up8x9-minmax-psimd.c",
     "src/f32-gavgpool/7p7x-psimd-c4.c",
     "src/f32-gavgpool/7x-psimd-c4.c",
-    "src/f32-gemm/gen/1x8-psimd-loadsplat.c",
-    "src/f32-gemm/gen/1x8-psimd-splat.c",
-    "src/f32-gemm/gen/1x8s4-psimd.c",
-    "src/f32-gemm/gen/4x2c4-psimd.c",
-    "src/f32-gemm/gen/4x8-psimd-loadsplat.c",
-    "src/f32-gemm/gen/4x8-psimd-splat.c",
-    "src/f32-gemm/gen/4x8s4-psimd.c",
-    "src/f32-gemm/gen/6x8-psimd-loadsplat.c",
-    "src/f32-gemm/gen/6x8-psimd-splat.c",
-    "src/f32-gemm/gen/6x8s4-psimd.c",
-    "src/f32-gemm/gen-inc/1x8-psimd-loadsplat.c",
-    "src/f32-gemm/gen-inc/1x8-psimd-splat.c",
-    "src/f32-gemm/gen-inc/1x8s4-psimd.c",
-    "src/f32-gemm/gen-inc/4x8-psimd-loadsplat.c",
-    "src/f32-gemm/gen-inc/4x8-psimd-splat.c",
-    "src/f32-gemm/gen-inc/4x8s4-psimd.c",
-    "src/f32-gemm/gen-inc/6x8-psimd-loadsplat.c",
-    "src/f32-gemm/gen-inc/6x8-psimd-splat.c",
-    "src/f32-gemm/gen-inc/6x8s4-psimd.c",
+    "src/f32-gemm/gen/1x8-minmax-psimd-loadsplat.c",
+    "src/f32-gemm/gen/1x8-minmax-psimd-splat.c",
+    "src/f32-gemm/gen/1x8s4-minmax-psimd.c",
+    "src/f32-gemm/gen/4x2c4-minmax-psimd.c",
+    "src/f32-gemm/gen/4x8-minmax-psimd-loadsplat.c",
+    "src/f32-gemm/gen/4x8-minmax-psimd-splat.c",
+    "src/f32-gemm/gen/4x8s4-minmax-psimd.c",
+    "src/f32-gemm/gen/6x8-minmax-psimd-loadsplat.c",
+    "src/f32-gemm/gen/6x8-minmax-psimd-splat.c",
+    "src/f32-gemm/gen/6x8s4-minmax-psimd.c",
+    "src/f32-gemm/gen-inc/1x8inc-minmax-psimd-loadsplat.c",
+    "src/f32-gemm/gen-inc/1x8inc-minmax-psimd-splat.c",
+    "src/f32-gemm/gen-inc/1x8s4inc-minmax-psimd.c",
+    "src/f32-gemm/gen-inc/4x8inc-minmax-psimd-loadsplat.c",
+    "src/f32-gemm/gen-inc/4x8inc-minmax-psimd-splat.c",
+    "src/f32-gemm/gen-inc/4x8s4inc-minmax-psimd.c",
+    "src/f32-gemm/gen-inc/6x8inc-minmax-psimd-loadsplat.c",
+    "src/f32-gemm/gen-inc/6x8inc-minmax-psimd-splat.c",
+    "src/f32-gemm/gen-inc/6x8s4inc-minmax-psimd.c",
     "src/f32-hswish/gen/psimd-x4.c",
     "src/f32-hswish/gen/psimd-x8.c",
     "src/f32-ibilinear/gen/psimd-c4.c",
     "src/f32-ibilinear/gen/psimd-c8.c",
-    "src/f32-igemm/gen/1x8-psimd-loadsplat.c",
-    "src/f32-igemm/gen/1x8-psimd-splat.c",
-    "src/f32-igemm/gen/1x8s4-psimd.c",
-    "src/f32-igemm/gen/4x2c4-psimd.c",
-    "src/f32-igemm/gen/4x8-psimd-loadsplat.c",
-    "src/f32-igemm/gen/4x8-psimd-splat.c",
-    "src/f32-igemm/gen/4x8s4-psimd.c",
-    "src/f32-igemm/gen/6x8-psimd-loadsplat.c",
-    "src/f32-igemm/gen/6x8-psimd-splat.c",
-    "src/f32-igemm/gen/6x8s4-psimd.c",
+    "src/f32-igemm/gen/1x8-minmax-psimd-loadsplat.c",
+    "src/f32-igemm/gen/1x8-minmax-psimd-splat.c",
+    "src/f32-igemm/gen/1x8s4-minmax-psimd.c",
+    "src/f32-igemm/gen/4x2c4-minmax-psimd.c",
+    "src/f32-igemm/gen/4x8-minmax-psimd-loadsplat.c",
+    "src/f32-igemm/gen/4x8-minmax-psimd-splat.c",
+    "src/f32-igemm/gen/4x8s4-minmax-psimd.c",
+    "src/f32-igemm/gen/6x8-minmax-psimd-loadsplat.c",
+    "src/f32-igemm/gen/6x8-minmax-psimd-splat.c",
+    "src/f32-igemm/gen/6x8s4-minmax-psimd.c",
     "src/f32-maxpool/9p8x-psimd-c4.c",
     "src/f32-pavgpool/9p8x-psimd-c4.c",
     "src/f32-pavgpool/9x-psimd-c4.c",
-    "src/f32-ppmm/gen/4x8-psimd.c",
+    "src/f32-ppmm/gen/4x8-minmax-psimd.c",
     "src/f32-prelu/gen/psimd-2x4.c",
     "src/f32-prelu/gen/psimd-2x8.c",
     "src/f32-rmax/psimd.c",
@@ -465,69 +465,69 @@
     "src/f32-avgpool/9x-neon-c4.c",
     "src/f32-clamp/gen/neon-x4.c",
     "src/f32-clamp/gen/neon-x8.c",
-    "src/f32-dwconv/gen/up4x9-neon.c",
-    "src/f32-dwconv/gen/up4x9-neon-acc2.c",
-    "src/f32-dwconv/gen/up8x9-neon.c",
-    "src/f32-dwconv/gen/up8x9-neon-acc2.c",
+    "src/f32-dwconv/gen/up4x9-minmax-neon.c",
+    "src/f32-dwconv/gen/up4x9-minmax-neon-acc2.c",
+    "src/f32-dwconv/gen/up8x9-minmax-neon.c",
+    "src/f32-dwconv/gen/up8x9-minmax-neon-acc2.c",
     "src/f32-gavgpool-spchw/neon-x4.c",
     "src/f32-gavgpool/7p7x-neon-c4.c",
     "src/f32-gavgpool/7x-neon-c4.c",
-    "src/f32-gemm/gen/1x8-neon-lane-ld64.c",
-    "src/f32-gemm/gen/4x2-neon-lane-ld64.c",
-    "src/f32-gemm/gen/4x8-neon-lane-ld128.c",
-    "src/f32-gemm/gen/4x8-neon-lane-ld64.c",
-    "src/f32-gemm/gen/5x8-neon-lane-ld64.c",
-    "src/f32-gemm/gen/6x8-neon-lane-ld64.c",
-    "src/f32-gemm/gen/6x8-neon-lane-ld128.c",
-    "src/f32-gemm/gen/1x8-neon-dup-ld64.c",
-    "src/f32-gemm/gen/4x8-neon-dup-ld128.c",
-    "src/f32-gemm/gen/4x8-neon-dup-ld64.c",
-    "src/f32-gemm/gen/6x8-neon-dup-ld64.c",
-    "src/f32-gemm/gen/6x8-neon-dup-ld128.c",
-    "src/f32-gemm/gen/1x8s4-neon.c",
-    "src/f32-gemm/gen/4x8s4-neon.c",
-    "src/f32-gemm/gen/6x8s4-neon.c",
-    "src/f32-gemm/gen/8x8s4-neon.c",
-    "src/f32-gemm/gen-inc/1x8-neon-lane-ld64.c",
-    "src/f32-gemm/gen-inc/4x8-neon-lane-ld128.c",
-    "src/f32-gemm/gen-inc/4x8-neon-lane-ld64.c",
-    "src/f32-gemm/gen-inc/5x8-neon-lane-ld64.c",
-    "src/f32-gemm/gen-inc/6x8-neon-lane-ld64.c",
-    "src/f32-gemm/gen-inc/6x8-neon-lane-ld128.c",
-    "src/f32-gemm/gen-inc/1x8-neon-dup-ld64.c",
-    "src/f32-gemm/gen-inc/4x8-neon-dup-ld128.c",
-    "src/f32-gemm/gen-inc/4x8-neon-dup-ld64.c",
-    "src/f32-gemm/gen-inc/6x8-neon-dup-ld64.c",
-    "src/f32-gemm/gen-inc/6x8-neon-dup-ld128.c",
-    "src/f32-gemm/gen-inc/1x8s4-neon.c",
-    "src/f32-gemm/gen-inc/4x8s4-neon.c",
-    "src/f32-gemm/gen-inc/6x8s4-neon.c",
-    "src/f32-gemm/gen-inc/8x8s4-neon.c",
+    "src/f32-gemm/gen/1x8-minmax-neon-lane-ld64.c",
+    "src/f32-gemm/gen/4x2-minmax-neon-lane-ld64.c",
+    "src/f32-gemm/gen/4x8-minmax-neon-lane-ld128.c",
+    "src/f32-gemm/gen/4x8-minmax-neon-lane-ld64.c",
+    "src/f32-gemm/gen/5x8-minmax-neon-lane-ld64.c",
+    "src/f32-gemm/gen/6x8-minmax-neon-lane-ld64.c",
+    "src/f32-gemm/gen/6x8-minmax-neon-lane-ld128.c",
+    "src/f32-gemm/gen/1x8-minmax-neon-dup-ld64.c",
+    "src/f32-gemm/gen/4x8-minmax-neon-dup-ld128.c",
+    "src/f32-gemm/gen/4x8-minmax-neon-dup-ld64.c",
+    "src/f32-gemm/gen/6x8-minmax-neon-dup-ld64.c",
+    "src/f32-gemm/gen/6x8-minmax-neon-dup-ld128.c",
+    "src/f32-gemm/gen/1x8s4-minmax-neon.c",
+    "src/f32-gemm/gen/4x8s4-minmax-neon.c",
+    "src/f32-gemm/gen/6x8s4-minmax-neon.c",
+    "src/f32-gemm/gen/8x8s4-minmax-neon.c",
+    "src/f32-gemm/gen-inc/1x8inc-minmax-neon-lane-ld64.c",
+    "src/f32-gemm/gen-inc/4x8inc-minmax-neon-lane-ld128.c",
+    "src/f32-gemm/gen-inc/4x8inc-minmax-neon-lane-ld64.c",
+    "src/f32-gemm/gen-inc/5x8inc-minmax-neon-lane-ld64.c",
+    "src/f32-gemm/gen-inc/6x8inc-minmax-neon-lane-ld64.c",
+    "src/f32-gemm/gen-inc/6x8inc-minmax-neon-lane-ld128.c",
+    "src/f32-gemm/gen-inc/1x8inc-minmax-neon-dup-ld64.c",
+    "src/f32-gemm/gen-inc/4x8inc-minmax-neon-dup-ld128.c",
+    "src/f32-gemm/gen-inc/4x8inc-minmax-neon-dup-ld64.c",
+    "src/f32-gemm/gen-inc/6x8inc-minmax-neon-dup-ld64.c",
+    "src/f32-gemm/gen-inc/6x8inc-minmax-neon-dup-ld128.c",
+    "src/f32-gemm/gen-inc/1x8s4inc-minmax-neon.c",
+    "src/f32-gemm/gen-inc/4x8s4inc-minmax-neon.c",
+    "src/f32-gemm/gen-inc/6x8s4inc-minmax-neon.c",
+    "src/f32-gemm/gen-inc/8x8s4inc-minmax-neon.c",
     "src/f32-hswish/gen/neon-x4.c",
     "src/f32-hswish/gen/neon-x8.c",
     "src/f32-ibilinear/gen/neon-c4.c",
     "src/f32-ibilinear/gen/neon-c8.c",
-    "src/f32-igemm/gen/1x8-neon-lane-ld64.c",
-    "src/f32-igemm/gen/4x2-neon-lane-ld64.c",
-    "src/f32-igemm/gen/4x4-neon-lane-ld64.c",
-    "src/f32-igemm/gen/4x8-neon-lane-ld128.c",
-    "src/f32-igemm/gen/4x8-neon-lane-ld64.c",
-    "src/f32-igemm/gen/6x8-neon-lane-ld64.c",
-    "src/f32-igemm/gen/6x8-neon-lane-ld128.c",
-    "src/f32-igemm/gen/1x8-neon-dup-ld64.c",
-    "src/f32-igemm/gen/4x8-neon-dup-ld128.c",
-    "src/f32-igemm/gen/4x8-neon-dup-ld64.c",
-    "src/f32-igemm/gen/6x8-neon-dup-ld64.c",
-    "src/f32-igemm/gen/6x8-neon-dup-ld128.c",
-    "src/f32-igemm/gen/1x8s4-neon.c",
-    "src/f32-igemm/gen/4x8s4-neon.c",
-    "src/f32-igemm/gen/6x8s4-neon.c",
-    "src/f32-igemm/gen/8x8s4-neon.c",
+    "src/f32-igemm/gen/1x8-minmax-neon-lane-ld64.c",
+    "src/f32-igemm/gen/4x2-minmax-neon-lane-ld64.c",
+    "src/f32-igemm/gen/4x4-minmax-neon-lane-ld64.c",
+    "src/f32-igemm/gen/4x8-minmax-neon-lane-ld128.c",
+    "src/f32-igemm/gen/4x8-minmax-neon-lane-ld64.c",
+    "src/f32-igemm/gen/6x8-minmax-neon-lane-ld64.c",
+    "src/f32-igemm/gen/6x8-minmax-neon-lane-ld128.c",
+    "src/f32-igemm/gen/1x8-minmax-neon-dup-ld64.c",
+    "src/f32-igemm/gen/4x8-minmax-neon-dup-ld128.c",
+    "src/f32-igemm/gen/4x8-minmax-neon-dup-ld64.c",
+    "src/f32-igemm/gen/6x8-minmax-neon-dup-ld64.c",
+    "src/f32-igemm/gen/6x8-minmax-neon-dup-ld128.c",
+    "src/f32-igemm/gen/1x8s4-minmax-neon.c",
+    "src/f32-igemm/gen/4x8s4-minmax-neon.c",
+    "src/f32-igemm/gen/6x8s4-minmax-neon.c",
+    "src/f32-igemm/gen/8x8s4-minmax-neon.c",
     "src/f32-maxpool/9p8x-neon-c4.c",
     "src/f32-pavgpool/9p8x-neon-c4.c",
     "src/f32-pavgpool/9x-neon-c4.c",
-    "src/f32-ppmm/gen/4x8-neon.c",
-    "src/f32-ppmm/gen/8x8-neon.c",
+    "src/f32-ppmm/gen/4x8-minmax-neon.c",
+    "src/f32-ppmm/gen/8x8-minmax-neon.c",
     "src/f32-prelu/gen/neon-2x4.c",
     "src/f32-prelu/gen/neon-2x8.c",
     "src/f32-raddstoreexpminusmax/gen/neon-p5-x4.c",
@@ -637,41 +637,41 @@
 NEONFMA_UKERNELS = [
     "src/f32-ibilinear/gen/neonfma-c4.c",
     "src/f32-ibilinear/gen/neonfma-c8.c",
-    "src/f32-igemm/gen/1x8-neonfma-dup-ld64.c",
-    "src/f32-igemm/gen/4x8-neonfma-dup-ld128.c",
-    "src/f32-igemm/gen/4x8-neonfma-dup-ld64.c",
-    "src/f32-igemm/gen/6x8-neonfma-dup-ld64.c",
-    "src/f32-igemm/gen/6x8-neonfma-dup-ld128.c",
-    "src/f32-igemm/gen/1x8s4-neonfma.c",
-    "src/f32-igemm/gen/4x8s4-neonfma.c",
-    "src/f32-igemm/gen/6x8s4-neonfma.c",
-    "src/f32-igemm/gen/8x8s4-neonfma.c",
-    "src/f32-dwconv/gen/up4x9-neonfma.c",
-    "src/f32-dwconv/gen/up4x9-neonfma-acc2.c",
-    "src/f32-dwconv/gen/up8x9-neonfma.c",
-    "src/f32-dwconv/gen/up8x9-neonfma-acc2.c",
-    "src/f32-gemm/gen/1x8-neonfma-dup-ld64.c",
-    "src/f32-gemm/gen/4x8-neonfma-dup-ld128.c",
-    "src/f32-gemm/gen/4x8-neonfma-dup-ld64.c",
-    "src/f32-gemm/gen/6x8-neonfma-dup-ld64.c",
-    "src/f32-gemm/gen/6x8-neonfma-dup-ld128.c",
-    "src/f32-gemm/gen/1x8s4-neonfma.c",
-    "src/f32-gemm/gen/4x8s4-neonfma.c",
-    "src/f32-gemm/gen/6x8s4-neonfma.c",
-    "src/f32-gemm/gen/8x8s4-neonfma.c",
-    "src/f32-gemm/gen-inc/1x8-neonfma-dup-ld64.c",
-    "src/f32-gemm/gen-inc/4x8-neonfma-dup-ld128.c",
-    "src/f32-gemm/gen-inc/4x8-neonfma-dup-ld64.c",
-    "src/f32-gemm/gen-inc/6x8-neonfma-dup-ld64.c",
-    "src/f32-gemm/gen-inc/6x8-neonfma-dup-ld128.c",
-    "src/f32-gemm/gen-inc/1x8s4-neonfma.c",
-    "src/f32-gemm/gen-inc/4x8s4-neonfma.c",
-    "src/f32-gemm/gen-inc/6x8s4-neonfma.c",
-    "src/f32-gemm/gen-inc/8x8s4-neonfma.c",
+    "src/f32-igemm/gen/1x8-minmax-neonfma-dup-ld64.c",
+    "src/f32-igemm/gen/4x8-minmax-neonfma-dup-ld128.c",
+    "src/f32-igemm/gen/4x8-minmax-neonfma-dup-ld64.c",
+    "src/f32-igemm/gen/6x8-minmax-neonfma-dup-ld64.c",
+    "src/f32-igemm/gen/6x8-minmax-neonfma-dup-ld128.c",
+    "src/f32-igemm/gen/1x8s4-minmax-neonfma.c",
+    "src/f32-igemm/gen/4x8s4-minmax-neonfma.c",
+    "src/f32-igemm/gen/6x8s4-minmax-neonfma.c",
+    "src/f32-igemm/gen/8x8s4-minmax-neonfma.c",
+    "src/f32-dwconv/gen/up4x9-minmax-neonfma.c",
+    "src/f32-dwconv/gen/up4x9-minmax-neonfma-acc2.c",
+    "src/f32-dwconv/gen/up8x9-minmax-neonfma.c",
+    "src/f32-dwconv/gen/up8x9-minmax-neonfma-acc2.c",
+    "src/f32-gemm/gen/1x8-minmax-neonfma-dup-ld64.c",
+    "src/f32-gemm/gen/4x8-minmax-neonfma-dup-ld128.c",
+    "src/f32-gemm/gen/4x8-minmax-neonfma-dup-ld64.c",
+    "src/f32-gemm/gen/6x8-minmax-neonfma-dup-ld64.c",
+    "src/f32-gemm/gen/6x8-minmax-neonfma-dup-ld128.c",
+    "src/f32-gemm/gen/1x8s4-minmax-neonfma.c",
+    "src/f32-gemm/gen/4x8s4-minmax-neonfma.c",
+    "src/f32-gemm/gen/6x8s4-minmax-neonfma.c",
+    "src/f32-gemm/gen/8x8s4-minmax-neonfma.c",
+    "src/f32-gemm/gen-inc/1x8inc-minmax-neonfma-dup-ld64.c",
+    "src/f32-gemm/gen-inc/4x8inc-minmax-neonfma-dup-ld128.c",
+    "src/f32-gemm/gen-inc/4x8inc-minmax-neonfma-dup-ld64.c",
+    "src/f32-gemm/gen-inc/6x8inc-minmax-neonfma-dup-ld64.c",
+    "src/f32-gemm/gen-inc/6x8inc-minmax-neonfma-dup-ld128.c",
+    "src/f32-gemm/gen-inc/1x8s4inc-minmax-neonfma.c",
+    "src/f32-gemm/gen-inc/4x8s4inc-minmax-neonfma.c",
+    "src/f32-gemm/gen-inc/6x8s4inc-minmax-neonfma.c",
+    "src/f32-gemm/gen-inc/8x8s4inc-minmax-neonfma.c",
     "src/f32-hswish/gen/neonfma-x4.c",
     "src/f32-hswish/gen/neonfma-x8.c",
-    "src/f32-ppmm/gen/4x8-neonfma.c",
-    "src/f32-ppmm/gen/8x8-neonfma.c",
+    "src/f32-ppmm/gen/4x8-minmax-neonfma.c",
+    "src/f32-ppmm/gen/8x8-minmax-neonfma.c",
     "src/f32-raddstoreexpminusmax/gen/neonfma-p5-x4.c",
     "src/f32-raddstoreexpminusmax/gen/neonfma-p5-x8.c",
     "src/f32-raddstoreexpminusmax/gen/neonfma-p5-x8-acc2.c",
@@ -784,26 +784,26 @@
     "src/f32-vbinary/gen/vdivc-neon-x8.c",
     "src/f32-vbinary/gen/vrdivc-neon-x4.c",
     "src/f32-vbinary/gen/vrdivc-neon-x8.c",
-    "src/f32-gemm/gen/1x8-neonfma-lane-ld64.c",
-    "src/f32-gemm/gen/4x2-neonfma-lane-ld64.c",
-    "src/f32-gemm/gen/4x8-neonfma-lane-ld128.c",
-    "src/f32-gemm/gen/4x8-neonfma-lane-ld64.c",
-    "src/f32-gemm/gen/5x8-neonfma-lane-ld64.c",
-    "src/f32-gemm/gen/6x8-neonfma-lane-ld64.c",
-    "src/f32-gemm/gen/6x8-neonfma-lane-ld128.c",
-    "src/f32-gemm/gen-inc/1x8-neonfma-lane-ld64.c",
-    "src/f32-gemm/gen-inc/4x8-neonfma-lane-ld128.c",
-    "src/f32-gemm/gen-inc/4x8-neonfma-lane-ld64.c",
-    "src/f32-gemm/gen-inc/5x8-neonfma-lane-ld64.c",
-    "src/f32-gemm/gen-inc/6x8-neonfma-lane-ld64.c",
-    "src/f32-gemm/gen-inc/6x8-neonfma-lane-ld128.c",
-    "src/f32-igemm/gen/1x8-neonfma-lane-ld64.c",
-    "src/f32-igemm/gen/4x2-neonfma-lane-ld64.c",
-    "src/f32-igemm/gen/4x4-neonfma-lane-ld64.c",
-    "src/f32-igemm/gen/4x8-neonfma-lane-ld128.c",
-    "src/f32-igemm/gen/4x8-neonfma-lane-ld64.c",
-    "src/f32-igemm/gen/6x8-neonfma-lane-ld64.c",
-    "src/f32-igemm/gen/6x8-neonfma-lane-ld128.c",
+    "src/f32-gemm/gen/1x8-minmax-neonfma-lane-ld64.c",
+    "src/f32-gemm/gen/4x2-minmax-neonfma-lane-ld64.c",
+    "src/f32-gemm/gen/4x8-minmax-neonfma-lane-ld128.c",
+    "src/f32-gemm/gen/4x8-minmax-neonfma-lane-ld64.c",
+    "src/f32-gemm/gen/5x8-minmax-neonfma-lane-ld64.c",
+    "src/f32-gemm/gen/6x8-minmax-neonfma-lane-ld64.c",
+    "src/f32-gemm/gen/6x8-minmax-neonfma-lane-ld128.c",
+    "src/f32-gemm/gen-inc/1x8inc-minmax-neonfma-lane-ld64.c",
+    "src/f32-gemm/gen-inc/4x8inc-minmax-neonfma-lane-ld128.c",
+    "src/f32-gemm/gen-inc/4x8inc-minmax-neonfma-lane-ld64.c",
+    "src/f32-gemm/gen-inc/5x8inc-minmax-neonfma-lane-ld64.c",
+    "src/f32-gemm/gen-inc/6x8inc-minmax-neonfma-lane-ld64.c",
+    "src/f32-gemm/gen-inc/6x8inc-minmax-neonfma-lane-ld128.c",
+    "src/f32-igemm/gen/1x8-minmax-neonfma-lane-ld64.c",
+    "src/f32-igemm/gen/4x2-minmax-neonfma-lane-ld64.c",
+    "src/f32-igemm/gen/4x4-minmax-neonfma-lane-ld64.c",
+    "src/f32-igemm/gen/4x8-minmax-neonfma-lane-ld128.c",
+    "src/f32-igemm/gen/4x8-minmax-neonfma-lane-ld64.c",
+    "src/f32-igemm/gen/6x8-minmax-neonfma-lane-ld64.c",
+    "src/f32-igemm/gen/6x8-minmax-neonfma-lane-ld128.c",
     "src/f32-conv-hwc/3x3s2p1c3x4-neonfma-2x2.c",
     "src/f32-conv-hwc/3x3s2p1c3x8-neonfma-2x2.c",
     "src/f32-conv-hwc2spchw/3x3s2p1c3x4-neonfma-2x2.c",
@@ -876,49 +876,49 @@
     "src/f32-clamp/gen/sse-x8.c",
     "src/f32-dwconv-spchw/3x3p1-sse.c",
     "src/f32-dwconv-spchw/3x3s2p1-sse.c",
-    "src/f32-dwconv/gen/up4x25-sse-acc2.c",
-    "src/f32-dwconv/gen/up4x25-sse.c",
-    "src/f32-dwconv/gen/up4x4-sse-acc2.c",
-    "src/f32-dwconv/gen/up4x4-sse.c",
-    "src/f32-dwconv/gen/up4x9-sse-acc2.c",
-    "src/f32-dwconv/gen/up4x9-sse.c",
-    "src/f32-dwconv/gen/up8x25-sse-acc2.c",
-    "src/f32-dwconv/gen/up8x25-sse.c",
-    "src/f32-dwconv/gen/up8x4-sse-acc2.c",
-    "src/f32-dwconv/gen/up8x4-sse.c",
-    "src/f32-dwconv/gen/up8x9-sse-acc2.c",
-    "src/f32-dwconv/gen/up8x9-sse.c",
+    "src/f32-dwconv/gen/up4x25-minmax-sse-acc2.c",
+    "src/f32-dwconv/gen/up4x25-minmax-sse.c",
+    "src/f32-dwconv/gen/up4x4-minmax-sse-acc2.c",
+    "src/f32-dwconv/gen/up4x4-minmax-sse.c",
+    "src/f32-dwconv/gen/up4x9-minmax-sse-acc2.c",
+    "src/f32-dwconv/gen/up4x9-minmax-sse.c",
+    "src/f32-dwconv/gen/up8x25-minmax-sse-acc2.c",
+    "src/f32-dwconv/gen/up8x25-minmax-sse.c",
+    "src/f32-dwconv/gen/up8x4-minmax-sse-acc2.c",
+    "src/f32-dwconv/gen/up8x4-minmax-sse.c",
+    "src/f32-dwconv/gen/up8x9-minmax-sse-acc2.c",
+    "src/f32-dwconv/gen/up8x9-minmax-sse.c",
     "src/f32-gavgpool-spchw/sse-x4.c",
     "src/f32-gavgpool/7p7x-sse-c4.c",
     "src/f32-gavgpool/7x-sse-c4.c",
-    "src/f32-gemm/gen/1x8-sse-dup.c",
-    "src/f32-gemm/gen/1x8-sse-load1.c",
-    "src/f32-gemm/gen/1x8s4-sse.c",
-    "src/f32-gemm/gen/4x2c4-sse.c",
-    "src/f32-gemm/gen/4x8-sse-dup.c",
-    "src/f32-gemm/gen/4x8-sse-load1.c",
-    "src/f32-gemm/gen/4x8s4-sse.c",
-    "src/f32-gemm/gen-inc/1x8-sse-dup.c",
-    "src/f32-gemm/gen-inc/1x8-sse-load1.c",
-    "src/f32-gemm/gen-inc/1x8s4-sse.c",
-    "src/f32-gemm/gen-inc/4x8-sse-dup.c",
-    "src/f32-gemm/gen-inc/4x8-sse-load1.c",
-    "src/f32-gemm/gen-inc/4x8s4-sse.c",
+    "src/f32-gemm/gen/1x8-minmax-sse-dup.c",
+    "src/f32-gemm/gen/1x8-minmax-sse-load1.c",
+    "src/f32-gemm/gen/1x8s4-minmax-sse.c",
+    "src/f32-gemm/gen/4x2c4-minmax-sse.c",
+    "src/f32-gemm/gen/4x8-minmax-sse-dup.c",
+    "src/f32-gemm/gen/4x8-minmax-sse-load1.c",
+    "src/f32-gemm/gen/4x8s4-minmax-sse.c",
+    "src/f32-gemm/gen-inc/1x8inc-minmax-sse-dup.c",
+    "src/f32-gemm/gen-inc/1x8inc-minmax-sse-load1.c",
+    "src/f32-gemm/gen-inc/1x8s4inc-minmax-sse.c",
+    "src/f32-gemm/gen-inc/4x8inc-minmax-sse-dup.c",
+    "src/f32-gemm/gen-inc/4x8inc-minmax-sse-load1.c",
+    "src/f32-gemm/gen-inc/4x8s4inc-minmax-sse.c",
     "src/f32-hswish/gen/sse-x4.c",
     "src/f32-hswish/gen/sse-x8.c",
     "src/f32-ibilinear/gen/sse-c4.c",
     "src/f32-ibilinear/gen/sse-c8.c",
-    "src/f32-igemm/gen/1x8-sse-dup.c",
-    "src/f32-igemm/gen/1x8-sse-load1.c",
-    "src/f32-igemm/gen/1x8s4-sse.c",
-    "src/f32-igemm/gen/4x2c4-sse.c",
-    "src/f32-igemm/gen/4x8-sse-dup.c",
-    "src/f32-igemm/gen/4x8-sse-load1.c",
-    "src/f32-igemm/gen/4x8s4-sse.c",
+    "src/f32-igemm/gen/1x8-minmax-sse-dup.c",
+    "src/f32-igemm/gen/1x8-minmax-sse-load1.c",
+    "src/f32-igemm/gen/1x8s4-minmax-sse.c",
+    "src/f32-igemm/gen/4x2c4-minmax-sse.c",
+    "src/f32-igemm/gen/4x8-minmax-sse-dup.c",
+    "src/f32-igemm/gen/4x8-minmax-sse-load1.c",
+    "src/f32-igemm/gen/4x8s4-minmax-sse.c",
     "src/f32-maxpool/9p8x-sse-c4.c",
     "src/f32-pavgpool/9p8x-sse-c4.c",
     "src/f32-pavgpool/9x-sse-c4.c",
-    "src/f32-ppmm/gen/4x8-sse.c",
+    "src/f32-ppmm/gen/4x8-minmax-sse.c",
     "src/f32-rmax/sse.c",
     "src/f32-spmm/gen/4x1-sse.c",
     "src/f32-spmm/gen/8x1-sse.c",
@@ -1032,47 +1032,47 @@
 AVX_UKERNELS = [
     "src/f32-clamp/gen/avx-x8.c",
     "src/f32-clamp/gen/avx-x16.c",
-    "src/f32-dwconv/gen/up16x4-avx-acc2.c",
-    "src/f32-dwconv/gen/up16x4-avx.c",
-    "src/f32-dwconv/gen/up8x4-avx-acc2.c",
-    "src/f32-dwconv/gen/up8x4-avx.c",
-    "src/f32-dwconv/gen/up16x9-avx-acc2.c",
-    "src/f32-dwconv/gen/up16x9-avx.c",
-    "src/f32-dwconv/gen/up8x9-avx-acc2.c",
-    "src/f32-dwconv/gen/up8x9-avx.c",
-    "src/f32-dwconv/gen/up16x25-avx-acc2.c",
-    "src/f32-dwconv/gen/up16x25-avx.c",
-    "src/f32-dwconv/gen/up8x25-avx-acc2.c",
-    "src/f32-dwconv/gen/up8x25-avx.c",
-    "src/f32-gemm/gen/1x8-avx-broadcast.c",
-    "src/f32-gemm/gen/4x8-avx-broadcast.c",
-    "src/f32-gemm/gen/5x8-avx-broadcast.c",
-    "src/f32-gemm/gen/6x8-avx-broadcast.c",
-    "src/f32-gemm/gen/7x8-avx-broadcast.c",
-    "src/f32-gemm/gen/1x16-avx-broadcast.c",
-    "src/f32-gemm/gen/3x16-avx-broadcast.c",
-    "src/f32-gemm/gen/4x16-avx-broadcast.c",
-    "src/f32-gemm/gen/5x16-avx-broadcast.c",
-    "src/f32-gemm/gen-inc/1x8-avx-broadcast.c",
-    "src/f32-gemm/gen-inc/4x8-avx-broadcast.c",
-    "src/f32-gemm/gen-inc/5x8-avx-broadcast.c",
-    "src/f32-gemm/gen-inc/6x8-avx-broadcast.c",
-    "src/f32-gemm/gen-inc/7x8-avx-broadcast.c",
-    "src/f32-gemm/gen-inc/1x16-avx-broadcast.c",
-    "src/f32-gemm/gen-inc/3x16-avx-broadcast.c",
-    "src/f32-gemm/gen-inc/4x16-avx-broadcast.c",
-    "src/f32-gemm/gen-inc/5x16-avx-broadcast.c",
+    "src/f32-dwconv/gen/up16x4-minmax-avx-acc2.c",
+    "src/f32-dwconv/gen/up16x4-minmax-avx.c",
+    "src/f32-dwconv/gen/up8x4-minmax-avx-acc2.c",
+    "src/f32-dwconv/gen/up8x4-minmax-avx.c",
+    "src/f32-dwconv/gen/up16x9-minmax-avx-acc2.c",
+    "src/f32-dwconv/gen/up16x9-minmax-avx.c",
+    "src/f32-dwconv/gen/up8x9-minmax-avx-acc2.c",
+    "src/f32-dwconv/gen/up8x9-minmax-avx.c",
+    "src/f32-dwconv/gen/up16x25-minmax-avx-acc2.c",
+    "src/f32-dwconv/gen/up16x25-minmax-avx.c",
+    "src/f32-dwconv/gen/up8x25-minmax-avx-acc2.c",
+    "src/f32-dwconv/gen/up8x25-minmax-avx.c",
+    "src/f32-gemm/gen/1x8-minmax-avx-broadcast.c",
+    "src/f32-gemm/gen/4x8-minmax-avx-broadcast.c",
+    "src/f32-gemm/gen/5x8-minmax-avx-broadcast.c",
+    "src/f32-gemm/gen/6x8-minmax-avx-broadcast.c",
+    "src/f32-gemm/gen/7x8-minmax-avx-broadcast.c",
+    "src/f32-gemm/gen/1x16-minmax-avx-broadcast.c",
+    "src/f32-gemm/gen/3x16-minmax-avx-broadcast.c",
+    "src/f32-gemm/gen/4x16-minmax-avx-broadcast.c",
+    "src/f32-gemm/gen/5x16-minmax-avx-broadcast.c",
+    "src/f32-gemm/gen-inc/1x8inc-minmax-avx-broadcast.c",
+    "src/f32-gemm/gen-inc/4x8inc-minmax-avx-broadcast.c",
+    "src/f32-gemm/gen-inc/5x8inc-minmax-avx-broadcast.c",
+    "src/f32-gemm/gen-inc/6x8inc-minmax-avx-broadcast.c",
+    "src/f32-gemm/gen-inc/7x8inc-minmax-avx-broadcast.c",
+    "src/f32-gemm/gen-inc/1x16inc-minmax-avx-broadcast.c",
+    "src/f32-gemm/gen-inc/3x16inc-minmax-avx-broadcast.c",
+    "src/f32-gemm/gen-inc/4x16inc-minmax-avx-broadcast.c",
+    "src/f32-gemm/gen-inc/5x16inc-minmax-avx-broadcast.c",
     "src/f32-hswish/gen/avx-x8.c",
     "src/f32-hswish/gen/avx-x16.c",
-    "src/f32-igemm/gen/1x8-avx-broadcast.c",
-    "src/f32-igemm/gen/4x8-avx-broadcast.c",
-    "src/f32-igemm/gen/5x8-avx-broadcast.c",
-    "src/f32-igemm/gen/6x8-avx-broadcast.c",
-    "src/f32-igemm/gen/7x8-avx-broadcast.c",
-    "src/f32-igemm/gen/1x16-avx-broadcast.c",
-    "src/f32-igemm/gen/3x16-avx-broadcast.c",
-    "src/f32-igemm/gen/4x16-avx-broadcast.c",
-    "src/f32-igemm/gen/5x16-avx-broadcast.c",
+    "src/f32-igemm/gen/1x8-minmax-avx-broadcast.c",
+    "src/f32-igemm/gen/4x8-minmax-avx-broadcast.c",
+    "src/f32-igemm/gen/5x8-minmax-avx-broadcast.c",
+    "src/f32-igemm/gen/6x8-minmax-avx-broadcast.c",
+    "src/f32-igemm/gen/7x8-minmax-avx-broadcast.c",
+    "src/f32-igemm/gen/1x16-minmax-avx-broadcast.c",
+    "src/f32-igemm/gen/3x16-minmax-avx-broadcast.c",
+    "src/f32-igemm/gen/4x16-minmax-avx-broadcast.c",
+    "src/f32-igemm/gen/5x16-minmax-avx-broadcast.c",
     "src/f32-prelu/gen/avx-2x8.c",
     "src/f32-prelu/gen/avx-2x16.c",
     "src/f32-rmax/avx.c",
@@ -1108,62 +1108,62 @@
 ]
 
 FMA3_UKERNELS = [
-    "src/f32-dwconv/gen/up16x4-fma3-acc2.c",
-    "src/f32-dwconv/gen/up16x4-fma3.c",
-    "src/f32-dwconv/gen/up8x4-fma3-acc2.c",
-    "src/f32-dwconv/gen/up8x4-fma3.c",
-    "src/f32-dwconv/gen/up16x9-fma3-acc2.c",
-    "src/f32-dwconv/gen/up16x9-fma3.c",
-    "src/f32-dwconv/gen/up8x9-fma3-acc2.c",
-    "src/f32-dwconv/gen/up8x9-fma3.c",
-    "src/f32-dwconv/gen/up16x25-fma3-acc2.c",
-    "src/f32-dwconv/gen/up16x25-fma3.c",
-    "src/f32-dwconv/gen/up8x25-fma3-acc2.c",
-    "src/f32-dwconv/gen/up8x25-fma3.c",
-    "src/f32-gemm/gen/1x8-fma3-broadcast.c",
-    "src/f32-gemm/gen/4x8-fma3-broadcast.c",
-    "src/f32-gemm/gen/5x8-fma3-broadcast.c",
-    "src/f32-gemm/gen/6x8-fma3-broadcast.c",
-    "src/f32-gemm/gen/7x8-fma3-broadcast.c",
-    "src/f32-gemm/gen/8x8-fma3-broadcast.c",
-    "src/f32-gemm/gen/1x16-fma3-broadcast.c",
-    "src/f32-gemm/gen/3x16-fma3-broadcast.c",
-    "src/f32-gemm/gen/4x16-fma3-broadcast.c",
-    "src/f32-gemm/gen/5x16-fma3-broadcast.c",
-    "src/f32-gemm/gen/1x16s4-fma3-broadcast.c",
-    "src/f32-gemm/gen/3x16s4-fma3-broadcast.c",
-    "src/f32-gemm/gen/4x16s4-fma3-broadcast.c",
-    "src/f32-gemm/gen/5x16s4-fma3-broadcast.c",
-    "src/f32-gemm/gen-inc/1x8-fma3-broadcast.c",
-    "src/f32-gemm/gen-inc/4x8-fma3-broadcast.c",
-    "src/f32-gemm/gen-inc/5x8-fma3-broadcast.c",
-    "src/f32-gemm/gen-inc/6x8-fma3-broadcast.c",
-    "src/f32-gemm/gen-inc/7x8-fma3-broadcast.c",
-    "src/f32-gemm/gen-inc/8x8-fma3-broadcast.c",
-    "src/f32-gemm/gen-inc/1x16-fma3-broadcast.c",
-    "src/f32-gemm/gen-inc/3x16-fma3-broadcast.c",
-    "src/f32-gemm/gen-inc/4x16-fma3-broadcast.c",
-    "src/f32-gemm/gen-inc/5x16-fma3-broadcast.c",
-    "src/f32-gemm/gen-inc/1x16s4-fma3-broadcast.c",
-    "src/f32-gemm/gen-inc/3x16s4-fma3-broadcast.c",
-    "src/f32-gemm/gen-inc/4x16s4-fma3-broadcast.c",
-    "src/f32-gemm/gen-inc/5x16s4-fma3-broadcast.c",
+    "src/f32-dwconv/gen/up16x4-minmax-fma3-acc2.c",
+    "src/f32-dwconv/gen/up16x4-minmax-fma3.c",
+    "src/f32-dwconv/gen/up8x4-minmax-fma3-acc2.c",
+    "src/f32-dwconv/gen/up8x4-minmax-fma3.c",
+    "src/f32-dwconv/gen/up16x9-minmax-fma3-acc2.c",
+    "src/f32-dwconv/gen/up16x9-minmax-fma3.c",
+    "src/f32-dwconv/gen/up8x9-minmax-fma3-acc2.c",
+    "src/f32-dwconv/gen/up8x9-minmax-fma3.c",
+    "src/f32-dwconv/gen/up16x25-minmax-fma3-acc2.c",
+    "src/f32-dwconv/gen/up16x25-minmax-fma3.c",
+    "src/f32-dwconv/gen/up8x25-minmax-fma3-acc2.c",
+    "src/f32-dwconv/gen/up8x25-minmax-fma3.c",
+    "src/f32-gemm/gen/1x8-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen/4x8-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen/5x8-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen/6x8-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen/7x8-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen/8x8-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen/1x16-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen/3x16-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen/4x16-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen/5x16-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen/1x16s4-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen/3x16s4-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen/4x16s4-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen/5x16s4-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen-inc/1x8inc-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen-inc/4x8inc-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen-inc/5x8inc-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen-inc/6x8inc-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen-inc/7x8inc-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen-inc/8x8inc-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen-inc/1x16inc-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen-inc/3x16inc-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen-inc/4x16inc-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen-inc/5x16inc-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen-inc/1x16s4inc-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen-inc/3x16s4inc-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen-inc/4x16s4inc-minmax-fma3-broadcast.c",
+    "src/f32-gemm/gen-inc/5x16s4inc-minmax-fma3-broadcast.c",
     "src/f32-hswish/gen/fma3-x8.c",
     "src/f32-hswish/gen/fma3-x16.c",
-    "src/f32-igemm/gen/1x8-fma3-broadcast.c",
-    "src/f32-igemm/gen/4x8-fma3-broadcast.c",
-    "src/f32-igemm/gen/5x8-fma3-broadcast.c",
-    "src/f32-igemm/gen/6x8-fma3-broadcast.c",
-    "src/f32-igemm/gen/7x8-fma3-broadcast.c",
-    "src/f32-igemm/gen/8x8-fma3-broadcast.c",
-    "src/f32-igemm/gen/1x16-fma3-broadcast.c",
-    "src/f32-igemm/gen/3x16-fma3-broadcast.c",
-    "src/f32-igemm/gen/4x16-fma3-broadcast.c",
-    "src/f32-igemm/gen/5x16-fma3-broadcast.c",
-    "src/f32-igemm/gen/1x16s4-fma3-broadcast.c",
-    "src/f32-igemm/gen/3x16s4-fma3-broadcast.c",
-    "src/f32-igemm/gen/4x16s4-fma3-broadcast.c",
-    "src/f32-igemm/gen/5x16s4-fma3-broadcast.c",
+    "src/f32-igemm/gen/1x8-minmax-fma3-broadcast.c",
+    "src/f32-igemm/gen/4x8-minmax-fma3-broadcast.c",
+    "src/f32-igemm/gen/5x8-minmax-fma3-broadcast.c",
+    "src/f32-igemm/gen/6x8-minmax-fma3-broadcast.c",
+    "src/f32-igemm/gen/7x8-minmax-fma3-broadcast.c",
+    "src/f32-igemm/gen/8x8-minmax-fma3-broadcast.c",
+    "src/f32-igemm/gen/1x16-minmax-fma3-broadcast.c",
+    "src/f32-igemm/gen/3x16-minmax-fma3-broadcast.c",
+    "src/f32-igemm/gen/4x16-minmax-fma3-broadcast.c",
+    "src/f32-igemm/gen/5x16-minmax-fma3-broadcast.c",
+    "src/f32-igemm/gen/1x16s4-minmax-fma3-broadcast.c",
+    "src/f32-igemm/gen/3x16s4-minmax-fma3-broadcast.c",
+    "src/f32-igemm/gen/4x16s4-minmax-fma3-broadcast.c",
+    "src/f32-igemm/gen/5x16s4-minmax-fma3-broadcast.c",
 ]
 
 AVX2_UKERNELS = [
@@ -1273,38 +1273,38 @@
 AVX512F_UKERNELS = [
     "src/f32-clamp/gen/avx512f-x16.c",
     "src/f32-clamp/gen/avx512f-x32.c",
-    "src/f32-dwconv/gen/up32x4-avx512f-acc2.c",
-    "src/f32-dwconv/gen/up32x4-avx512f.c",
-    "src/f32-dwconv/gen/up16x4-avx512f-acc2.c",
-    "src/f32-dwconv/gen/up16x4-avx512f.c",
-    "src/f32-dwconv/gen/up32x9-avx512f-acc2.c",
-    "src/f32-dwconv/gen/up32x9-avx512f.c",
-    "src/f32-dwconv/gen/up16x9-avx512f-acc2.c",
-    "src/f32-dwconv/gen/up16x9-avx512f.c",
-    "src/f32-dwconv/gen/up32x25-avx512f-acc2.c",
-    "src/f32-dwconv/gen/up32x25-avx512f.c",
-    "src/f32-dwconv/gen/up16x25-avx512f-acc2.c",
-    "src/f32-dwconv/gen/up16x25-avx512f.c",
-    "src/f32-gemm/gen/1x16-avx512f-broadcast.c",
-    "src/f32-gemm/gen/4x16-avx512f-broadcast.c",
-    "src/f32-gemm/gen/5x16-avx512f-broadcast.c",
-    "src/f32-gemm/gen/6x16-avx512f-broadcast.c",
-    "src/f32-gemm/gen/7x16-avx512f-broadcast.c",
-    "src/f32-gemm/gen/8x16-avx512f-broadcast.c",
-    "src/f32-gemm/gen-inc/1x16-avx512f-broadcast.c",
-    "src/f32-gemm/gen-inc/4x16-avx512f-broadcast.c",
-    "src/f32-gemm/gen-inc/5x16-avx512f-broadcast.c",
-    "src/f32-gemm/gen-inc/6x16-avx512f-broadcast.c",
-    "src/f32-gemm/gen-inc/7x16-avx512f-broadcast.c",
-    "src/f32-gemm/gen-inc/8x16-avx512f-broadcast.c",
+    "src/f32-dwconv/gen/up32x4-minmax-avx512f-acc2.c",
+    "src/f32-dwconv/gen/up32x4-minmax-avx512f.c",
+    "src/f32-dwconv/gen/up16x4-minmax-avx512f-acc2.c",
+    "src/f32-dwconv/gen/up16x4-minmax-avx512f.c",
+    "src/f32-dwconv/gen/up32x9-minmax-avx512f-acc2.c",
+    "src/f32-dwconv/gen/up32x9-minmax-avx512f.c",
+    "src/f32-dwconv/gen/up16x9-minmax-avx512f-acc2.c",
+    "src/f32-dwconv/gen/up16x9-minmax-avx512f.c",
+    "src/f32-dwconv/gen/up32x25-minmax-avx512f-acc2.c",
+    "src/f32-dwconv/gen/up32x25-minmax-avx512f.c",
+    "src/f32-dwconv/gen/up16x25-minmax-avx512f-acc2.c",
+    "src/f32-dwconv/gen/up16x25-minmax-avx512f.c",
+    "src/f32-gemm/gen/1x16-minmax-avx512f-broadcast.c",
+    "src/f32-gemm/gen/4x16-minmax-avx512f-broadcast.c",
+    "src/f32-gemm/gen/5x16-minmax-avx512f-broadcast.c",
+    "src/f32-gemm/gen/6x16-minmax-avx512f-broadcast.c",
+    "src/f32-gemm/gen/7x16-minmax-avx512f-broadcast.c",
+    "src/f32-gemm/gen/8x16-minmax-avx512f-broadcast.c",
+    "src/f32-gemm/gen-inc/1x16inc-minmax-avx512f-broadcast.c",
+    "src/f32-gemm/gen-inc/4x16inc-minmax-avx512f-broadcast.c",
+    "src/f32-gemm/gen-inc/5x16inc-minmax-avx512f-broadcast.c",
+    "src/f32-gemm/gen-inc/6x16inc-minmax-avx512f-broadcast.c",
+    "src/f32-gemm/gen-inc/7x16inc-minmax-avx512f-broadcast.c",
+    "src/f32-gemm/gen-inc/8x16inc-minmax-avx512f-broadcast.c",
     "src/f32-hswish/gen/avx512f-x16.c",
     "src/f32-hswish/gen/avx512f-x32.c",
-    "src/f32-igemm/gen/1x16-avx512f-broadcast.c",
-    "src/f32-igemm/gen/4x16-avx512f-broadcast.c",
-    "src/f32-igemm/gen/5x16-avx512f-broadcast.c",
-    "src/f32-igemm/gen/6x16-avx512f-broadcast.c",
-    "src/f32-igemm/gen/7x16-avx512f-broadcast.c",
-    "src/f32-igemm/gen/8x16-avx512f-broadcast.c",
+    "src/f32-igemm/gen/1x16-minmax-avx512f-broadcast.c",
+    "src/f32-igemm/gen/4x16-minmax-avx512f-broadcast.c",
+    "src/f32-igemm/gen/5x16-minmax-avx512f-broadcast.c",
+    "src/f32-igemm/gen/6x16-minmax-avx512f-broadcast.c",
+    "src/f32-igemm/gen/7x16-minmax-avx512f-broadcast.c",
+    "src/f32-igemm/gen/8x16-minmax-avx512f-broadcast.c",
     "src/f32-prelu/gen/avx512f-2x16.c",
     "src/f32-prelu/gen/avx512f-2x32.c",
     "src/f32-raddexpminusmax/gen/avx512f-p5-scalef-x128.c",
@@ -1406,80 +1406,80 @@
 
 AARCH32_ASM_UKERNELS = [
     "src/q8-dwconv/up8x9-aarch32-neon.S",
-    "src/f32-gemm/4x8-aarch32-neon-cortex-a53.S",
-    "src/f32-gemm/4x8-aarch32-neon-cortex-a55.S",
-    "src/f32-gemm/gen/4x8-aarch32-neon-cortex-a75.S",
-    "src/f32-gemm/gen/4x8-aarch32-neon-pld-cortex-a75.S",
-    "src/f32-gemm/4x8-aarch32-neon-ld64.S",
-    "src/f32-igemm/4x8-aarch32-neon-ld64.S",
-    "src/f32-igemm/gen/4x8-aarch32-neon-cortex-a75.S",
-    "src/f32-igemm/gen/4x8-aarch32-neon-pld-cortex-a75.S",
-    "src/f32-igemm/4x8-aarch32-neon-cortex-a53.S",
-    "src/f32-igemm/4x8-aarch32-neon-cortex-a55.S",
+    "src/f32-gemm/4x8-minmax-aarch32-neon-cortex-a53.S",
+    "src/f32-gemm/4x8-minmax-aarch32-neon-cortex-a55.S",
+    "src/f32-gemm/gen/4x8-minmax-aarch32-neon-cortex-a75.S",
+    "src/f32-gemm/gen/4x8-minmax-aarch32-neon-pld-cortex-a75.S",
+    "src/f32-gemm/4x8-minmax-aarch32-neon-ld64.S",
+    "src/f32-igemm/4x8-minmax-aarch32-neon-ld64.S",
+    "src/f32-igemm/gen/4x8-minmax-aarch32-neon-cortex-a75.S",
+    "src/f32-igemm/gen/4x8-minmax-aarch32-neon-pld-cortex-a75.S",
+    "src/f32-igemm/4x8-minmax-aarch32-neon-cortex-a53.S",
+    "src/f32-igemm/4x8-minmax-aarch32-neon-cortex-a55.S",
 ]
 
 AARCH64_ASM_UKERNELS = [
-    "src/f32-dwconv/up4x9-aarch64-neonfma-cortex-a55.S",
-    "src/f32-dwconv/up4x9-aarch64-neonfma.S",
-    "src/f32-gemm/gen/1x12-aarch64-neonfma-cortex-a53.S",
-    "src/f32-gemm/gen/1x8-aarch64-neonfma-cortex-a53.S",
-    "src/f32-gemm/gen/1x8-aarch64-neonfma-cortex-a57.S",
-    "src/f32-gemm/gen/1x8-aarch64-neonfma-cortex-a75.S",
-    "src/f32-gemm/gen/4x12-aarch64-neonfma-cortex-a53.S",
-    "src/f32-gemm/gen/4x8-aarch64-neonfma-cortex-a53.S",
-    "src/f32-gemm/gen/4x8-aarch64-neonfma-cortex-a55.S",
-    "src/f32-gemm/gen/4x8-aarch64-neonfma-cortex-a57.S",
-    "src/f32-gemm/gen/4x8-aarch64-neonfma-cortex-a75.S",
-    "src/f32-gemm/gen/4x8-aarch64-neonfma-ld128.S",
-    "src/f32-gemm/gen/4x8-aarch64-neonfma-ld64.S",
-    "src/f32-gemm/gen/5x8-aarch64-neonfma-cortex-a57.S",
-    "src/f32-gemm/gen/5x8-aarch64-neonfma-cortex-a75.S",
-    "src/f32-gemm/gen/6x8-aarch64-neonfma-cortex-a53.S",
-    "src/f32-gemm/gen/6x8-aarch64-neonfma-cortex-a55.S",
-    "src/f32-gemm/gen/6x8-aarch64-neonfma-cortex-a73.S",
-    "src/f32-gemm/gen/6x8-aarch64-neonfma-cortex-a57.S",
-    "src/f32-gemm/gen/6x8-aarch64-neonfma-cortex-a75.S",
-    "src/f32-gemm/gen/6x8-aarch64-neonfma-ios.S",
-    "src/f32-gemm/gen/6x8-aarch64-neonfma-ld128.S",
-    "src/f32-gemm/gen/6x8-aarch64-neonfma-ld64.S",
-    "src/f32-gemm/gen-inc/1x12-aarch64-neonfma-cortex-a53.S",
-    "src/f32-gemm/gen-inc/1x8-aarch64-neonfma-cortex-a53.S",
-    "src/f32-gemm/gen-inc/1x8-aarch64-neonfma-cortex-a57.S",
-    "src/f32-gemm/gen-inc/1x8-aarch64-neonfma-cortex-a75.S",
-    "src/f32-gemm/gen-inc/4x12-aarch64-neonfma-cortex-a53.S",
-    "src/f32-gemm/gen-inc/4x8-aarch64-neonfma-cortex-a53.S",
-    "src/f32-gemm/gen-inc/4x8-aarch64-neonfma-cortex-a55.S",
-    "src/f32-gemm/gen-inc/4x8-aarch64-neonfma-cortex-a57.S",
-    "src/f32-gemm/gen-inc/4x8-aarch64-neonfma-cortex-a75.S",
-    "src/f32-gemm/gen-inc/4x8-aarch64-neonfma-ld128.S",
-    "src/f32-gemm/gen-inc/4x8-aarch64-neonfma-ld64.S",
-    "src/f32-gemm/gen-inc/5x8-aarch64-neonfma-cortex-a57.S",
-    "src/f32-gemm/gen-inc/5x8-aarch64-neonfma-cortex-a75.S",
-    "src/f32-gemm/gen-inc/6x8-aarch64-neonfma-cortex-a53.S",
-    "src/f32-gemm/gen-inc/6x8-aarch64-neonfma-cortex-a55.S",
-    "src/f32-gemm/gen-inc/6x8-aarch64-neonfma-cortex-a73.S",
-    "src/f32-gemm/gen-inc/6x8-aarch64-neonfma-cortex-a57.S",
-    "src/f32-gemm/gen-inc/6x8-aarch64-neonfma-cortex-a75.S",
-    "src/f32-gemm/gen-inc/6x8-aarch64-neonfma-ios.S",
-    "src/f32-gemm/gen-inc/6x8-aarch64-neonfma-ld128.S",
-    "src/f32-gemm/gen-inc/6x8-aarch64-neonfma-ld64.S",
-    "src/f32-igemm/1x12-aarch64-neonfma-cortex-a53.S",
-    "src/f32-igemm/1x8-aarch64-neonfma-cortex-a53.S",
-    "src/f32-igemm/gen/1x8-aarch64-neonfma-cortex-a57.S",
-    "src/f32-igemm/gen/1x8-aarch64-neonfma-cortex-a75.S",
-    "src/f32-igemm/4x12-aarch64-neonfma-cortex-a53.S",
-    "src/f32-igemm/4x8-aarch64-neonfma-cortex-a53.S",
-    "src/f32-igemm/4x8-aarch64-neonfma-cortex-a55.S",
-    "src/f32-igemm/gen/4x8-aarch64-neonfma-cortex-a57.S",
-    "src/f32-igemm/gen/4x8-aarch64-neonfma-cortex-a75.S",
-    "src/f32-igemm/gen/5x8-aarch64-neonfma-cortex-a57.S",
-    "src/f32-igemm/gen/5x8-aarch64-neonfma-cortex-a75.S",
-    "src/f32-igemm/6x8-aarch64-neonfma-cortex-a53.S",
-    "src/f32-igemm/6x8-aarch64-neonfma-cortex-a55.S",
-    "src/f32-igemm/6x8-aarch64-neonfma-cortex-a73.S",
-    "src/f32-igemm/gen/6x8-aarch64-neonfma-cortex-a57.S",
-    "src/f32-igemm/gen/6x8-aarch64-neonfma-cortex-a75.S",
-    "src/f32-igemm/gen/6x8-aarch64-neonfma-ios.S",
+    "src/f32-dwconv/up4x9-minmax-aarch64-neonfma-cortex-a55.S",
+    "src/f32-dwconv/up4x9-minmax-aarch64-neonfma.S",
+    "src/f32-gemm/gen/1x12-minmax-aarch64-neonfma-cortex-a53.S",
+    "src/f32-gemm/gen/1x8-minmax-aarch64-neonfma-cortex-a53.S",
+    "src/f32-gemm/gen/1x8-minmax-aarch64-neonfma-cortex-a57.S",
+    "src/f32-gemm/gen/1x8-minmax-aarch64-neonfma-cortex-a75.S",
+    "src/f32-gemm/gen/4x12-minmax-aarch64-neonfma-cortex-a53.S",
+    "src/f32-gemm/gen/4x8-minmax-aarch64-neonfma-cortex-a53.S",
+    "src/f32-gemm/gen/4x8-minmax-aarch64-neonfma-cortex-a55.S",
+    "src/f32-gemm/gen/4x8-minmax-aarch64-neonfma-cortex-a57.S",
+    "src/f32-gemm/gen/4x8-minmax-aarch64-neonfma-cortex-a75.S",
+    "src/f32-gemm/gen/4x8-minmax-aarch64-neonfma-ld128.S",
+    "src/f32-gemm/gen/4x8-minmax-aarch64-neonfma-ld64.S",
+    "src/f32-gemm/gen/5x8-minmax-aarch64-neonfma-cortex-a57.S",
+    "src/f32-gemm/gen/5x8-minmax-aarch64-neonfma-cortex-a75.S",
+    "src/f32-gemm/gen/6x8-minmax-aarch64-neonfma-cortex-a53.S",
+    "src/f32-gemm/gen/6x8-minmax-aarch64-neonfma-cortex-a55.S",
+    "src/f32-gemm/gen/6x8-minmax-aarch64-neonfma-cortex-a73.S",
+    "src/f32-gemm/gen/6x8-minmax-aarch64-neonfma-cortex-a57.S",
+    "src/f32-gemm/gen/6x8-minmax-aarch64-neonfma-cortex-a75.S",
+    "src/f32-gemm/gen/6x8-minmax-aarch64-neonfma-ios.S",
+    "src/f32-gemm/gen/6x8-minmax-aarch64-neonfma-ld128.S",
+    "src/f32-gemm/gen/6x8-minmax-aarch64-neonfma-ld64.S",
+    "src/f32-gemm/gen-inc/1x12inc-minmax-aarch64-neonfma-cortex-a53.S",
+    "src/f32-gemm/gen-inc/1x8inc-minmax-aarch64-neonfma-cortex-a53.S",
+    "src/f32-gemm/gen-inc/1x8inc-minmax-aarch64-neonfma-cortex-a57.S",
+    "src/f32-gemm/gen-inc/1x8inc-minmax-aarch64-neonfma-cortex-a75.S",
+    "src/f32-gemm/gen-inc/4x12inc-minmax-aarch64-neonfma-cortex-a53.S",
+    "src/f32-gemm/gen-inc/4x8inc-minmax-aarch64-neonfma-cortex-a53.S",
+    "src/f32-gemm/gen-inc/4x8inc-minmax-aarch64-neonfma-cortex-a55.S",
+    "src/f32-gemm/gen-inc/4x8inc-minmax-aarch64-neonfma-cortex-a57.S",
+    "src/f32-gemm/gen-inc/4x8inc-minmax-aarch64-neonfma-cortex-a75.S",
+    "src/f32-gemm/gen-inc/4x8inc-minmax-aarch64-neonfma-ld128.S",
+    "src/f32-gemm/gen-inc/4x8inc-minmax-aarch64-neonfma-ld64.S",
+    "src/f32-gemm/gen-inc/5x8inc-minmax-aarch64-neonfma-cortex-a57.S",
+    "src/f32-gemm/gen-inc/5x8inc-minmax-aarch64-neonfma-cortex-a75.S",
+    "src/f32-gemm/gen-inc/6x8inc-minmax-aarch64-neonfma-cortex-a53.S",
+    "src/f32-gemm/gen-inc/6x8inc-minmax-aarch64-neonfma-cortex-a55.S",
+    "src/f32-gemm/gen-inc/6x8inc-minmax-aarch64-neonfma-cortex-a73.S",
+    "src/f32-gemm/gen-inc/6x8inc-minmax-aarch64-neonfma-cortex-a57.S",
+    "src/f32-gemm/gen-inc/6x8inc-minmax-aarch64-neonfma-cortex-a75.S",
+    "src/f32-gemm/gen-inc/6x8inc-minmax-aarch64-neonfma-ios.S",
+    "src/f32-gemm/gen-inc/6x8inc-minmax-aarch64-neonfma-ld128.S",
+    "src/f32-gemm/gen-inc/6x8inc-minmax-aarch64-neonfma-ld64.S",
+    "src/f32-igemm/1x12-minmax-aarch64-neonfma-cortex-a53.S",
+    "src/f32-igemm/1x8-minmax-aarch64-neonfma-cortex-a53.S",
+    "src/f32-igemm/gen/1x8-minmax-aarch64-neonfma-cortex-a57.S",
+    "src/f32-igemm/gen/1x8-minmax-aarch64-neonfma-cortex-a75.S",
+    "src/f32-igemm/4x12-minmax-aarch64-neonfma-cortex-a53.S",
+    "src/f32-igemm/4x8-minmax-aarch64-neonfma-cortex-a53.S",
+    "src/f32-igemm/4x8-minmax-aarch64-neonfma-cortex-a55.S",
+    "src/f32-igemm/gen/4x8-minmax-aarch64-neonfma-cortex-a57.S",
+    "src/f32-igemm/gen/4x8-minmax-aarch64-neonfma-cortex-a75.S",
+    "src/f32-igemm/gen/5x8-minmax-aarch64-neonfma-cortex-a57.S",
+    "src/f32-igemm/gen/5x8-minmax-aarch64-neonfma-cortex-a75.S",
+    "src/f32-igemm/6x8-minmax-aarch64-neonfma-cortex-a53.S",
+    "src/f32-igemm/6x8-minmax-aarch64-neonfma-cortex-a55.S",
+    "src/f32-igemm/6x8-minmax-aarch64-neonfma-cortex-a73.S",
+    "src/f32-igemm/gen/6x8-minmax-aarch64-neonfma-cortex-a57.S",
+    "src/f32-igemm/gen/6x8-minmax-aarch64-neonfma-cortex-a75.S",
+    "src/f32-igemm/gen/6x8-minmax-aarch64-neonfma-ios.S",
 ]
 
 INTERNAL_MICROKERNEL_HDRS = [
@@ -2471,9 +2471,9 @@
 )
 
 xnnpack_unit_test(
-    name = "f32_igemm_test",
+    name = "f32_igemm_minmax_test",
     srcs = [
-        "test/f32-igemm.cc",
+        "test/f32-igemm-minmax.cc",
         "test/gemm-microkernel-tester.h",
         "src/xnnpack/AlignedAllocator.h",
     ] + WEIGHTS_PACK_HDRS + MICROKERNEL_TEST_HDRS,
@@ -2501,9 +2501,9 @@
 )
 
 xnnpack_unit_test(
-    name = "f32_dwconv_test",
+    name = "f32_dwconv_minmax_test",
     srcs = [
-        "test/f32-dwconv.cc",
+        "test/f32-dwconv-minmax.cc",
         "test/dwconv-microkernel-tester.h",
         "src/xnnpack/AlignedAllocator.h",
     ] + WEIGHTS_PACK_HDRS + MICROKERNEL_TEST_HDRS,
@@ -2541,9 +2541,9 @@
 )
 
 xnnpack_unit_test(
-    name = "f32_gemm_test",
+    name = "f32_gemm_minmax_test",
     srcs = [
-        "test/f32-gemm.cc",
+        "test/f32-gemm-minmax.cc",
         "test/gemm-microkernel-tester.h",
         "src/xnnpack/AlignedAllocator.h",
     ] + WEIGHTS_PACK_HDRS + MICROKERNEL_TEST_HDRS,
@@ -2551,9 +2551,9 @@
 )
 
 xnnpack_unit_test(
-    name = "f32_gemminc_test",
+    name = "f32_gemminc_minmax_test",
     srcs = [
-        "test/f32-gemminc.cc",
+        "test/f32-gemminc-minmax.cc",
         "test/gemm-microkernel-tester.h",
         "src/xnnpack/AlignedAllocator.h",
     ] + WEIGHTS_PACK_HDRS + MICROKERNEL_TEST_HDRS,
@@ -2589,9 +2589,9 @@
 )
 
 xnnpack_unit_test(
-    name = "f32_ppmm_test",
+    name = "f32_ppmm_minmax_test",
     srcs = [
-        "test/f32-ppmm.cc",
+        "test/f32-ppmm-minmax.cc",
         "test/gemm-microkernel-tester.h",
         "src/xnnpack/AlignedAllocator.h",
     ] + WEIGHTS_PACK_HDRS + MICROKERNEL_TEST_HDRS,